Re: [Linuxptp-devel] [PATCH RFC 2/3] Inhibit sync messages from unwilling 802.1AS ports.

2013-12-03 Thread Richard Cochran
On Tue, Dec 03, 2013 at 11:39:30AM +0100, Delio Brignoli wrote: As would I. Thank you Richard for taking the time to prepare this patch series. So does this fix the issue you were seeing? If so, I'll respin it, adjusted for the comments. Thanks, Richard

Re: [Linuxptp-devel] [PATCH] sk: update error message for Tx timestamp polling

2013-12-04 Thread Richard Cochran
On Wed, Dec 04, 2013 at 02:35:51PM +0800, Arnold kang wrote: hi all: thease days, i just built ptp for linux arm, all seems goes right, but there a bug in sk.c, i cann't create the poll for Tx timestamp polling. i increase the sk_tx_timeout, but that not work. i'm trying other way, hope

Re: [Linuxptp-devel] ptp4l push notifications

2013-12-10 Thread Richard Cochran
On Mon, Dec 09, 2013 at 07:39:19PM +, Keller, Jacob E wrote: How would this work with multiple management interfaces? Is that even possible? IE: multiple programs connected? Does the management interface properly only send to a single client? If so we can use the client-request

Re: [Linuxptp-devel] [PATCH] Fix drift calculation in PI servo with large values.

2014-01-01 Thread Richard Cochran
On Fri, Dec 20, 2013 at 04:51:35PM +0100, Miroslav Lichvar wrote: When the drift value is adjusted by the newly measured frequency offset, multiply the frequencies instead of adding the measured offset to the old value to get accurate result even when updating a large drift. In other words,

Re: [Linuxptp-devel] [PATCH 0/3] Improve ioctl to request hardware features

2014-01-05 Thread Richard Cochran
On Sat, Jan 04, 2014 at 02:54:56PM -0200, Flavio Leitner wrote: I presume from what you are saying that broken hardware isn't that common and I might be just unlucky to get my hands on two. Yes, that is just what I wanted to say. All of the other cards for which there are Linux drivers are

[Linuxptp-devel] [PATCH RFC 1/2] uds: take the interface data off the stack.

2014-01-08 Thread Richard Cochran
part of the clock instance. Signed-off-by: Richard Cochran richardcoch...@gmail.com Reported-by: Miroslav Lichvar mlich...@redhat.com --- clock.c | 14 +++--- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/clock.c b/clock.c index 063155b..c99812b 100644 --- a/clock.c +++ b

[Linuxptp-devel] [PATCH RFC 0/2] UDS bug fixes

2014-01-08 Thread Richard Cochran
Here is how I propose to deal with the recent UDS bugs. Comments are most welcome. Thanks, Richard Richard Cochran (2): uds: take the interface data off the stack. uds: disable the accidentally enabled announce timer. clock.c | 14 +++--- ds.h|1 + port.c |3

[Linuxptp-devel] Release planning and looking ahead

2014-02-07 Thread Richard Cochran
It has been six months since the last release, and activity has been quiet for a while, so I plan to release v1.4 in about one week. If you want to do any last minute testing (or you want to report a bug), now is the time to do it. After that, we can start on adding new requested features. Here

Re: [Linuxptp-devel] [PATCH] port: reject negative peer delay in port_peer_delay()

2014-02-18 Thread Richard Cochran
On Wed, Feb 12, 2014 at 09:58:07PM +0100, dbrign...@audioscience.com wrote: From: Delio Brignoli dbrign...@audioscience.com Peer delay should never be negative, when it occurs warn the user and drop it. I don't think it is right to drop such measurements. If the clocks are far enough apart

Re: [Linuxptp-devel] [PATCH] port: reject negative peer delay in port_peer_delay()

2014-02-18 Thread Richard Cochran
On Wed, Feb 19, 2014 at 12:38:59AM +0100, Holzinger, Axel (ALC NetworX GmbH) wrote: So this means using the UDS Interface I guess. But does it really make sense to use UDS for reading a precise time. Is it fast enough? No, it is not for getting time stamps. For that, you read the PHC

Re: [Linuxptp-devel] [PATCH] port: reject negative peer delay in port_peer_delay()

2014-02-19 Thread Richard Cochran
On Tue, Feb 18, 2014 at 07:08:41PM +0100, Delio Brignoli wrote: I also assumed the peer delay had to be non-negative to be meaningful. Would you be OK with a patch that introduced a minimum acceptable peer delay configuration option? I think even for gPTP, the measured mean peer propagation

Re: [Linuxptp-devel] [PATCH v2 2/2] config: Add min_neighbor_prop_delay configuration variable

2014-02-21 Thread Richard Cochran
On Fri, Feb 21, 2014 at 10:40:55AM +0100, dbrign...@audioscience.com wrote: From: Delio Brignoli dbrign...@audioscience.com When peer delay is min_neighbor_prop_delay the port is flagged as non 802.1AS capable. min_neighbor_prop_delay defaults to INT_MIN. Signed-off-by: Delio Brignoli

Re: [Linuxptp-devel] [PATCH 00/27] automatic phc2sys configuration

2014-03-23 Thread Richard Cochran
On Thu, Mar 20, 2014 at 07:08:13PM +0100, Jiri Benc wrote: This patchset adds support for phc2sys following ptp4l state changes and automatic (re)configuration. This could be very useful indeed. The automatic configuration does not use PPS for now. With PPS, things get much more complicated;

Re: [Linuxptp-devel] [PATCH v2 03/27] Allow sending to a specified (unicast) address

2014-04-02 Thread Richard Cochran
On Mon, Mar 24, 2014 at 09:53:20AM +0100, Jiri Benc wrote: +/* + * Send message to the given address. This is intended for management + * messages only, thus goes always to the general port. There are a couple of unicast options for PTP. We should try to make this new interface in a way that

Re: [Linuxptp-devel] [PATCH v2 01/27] Move check of TLV length for management COMMAND messages

2014-04-05 Thread Richard Cochran
On Mon, Mar 24, 2014 at 09:53:18AM +0100, Jiri Benc wrote: Currently, it is assumed that the management TLV data of management COMMAND messages is always empty. This is not true for the INITIALIZE command and also for a custom command we'll be introducing. Move the check to msg_post_recv and

Re: [Linuxptp-devel] [PATCH v2 04/27] uds: don't output Connection refused

2014-04-05 Thread Richard Cochran
On Mon, Mar 24, 2014 at 09:53:21AM +0100, Jiri Benc wrote: When phc2sys is started before ptp4l or it is interrupted before ptp4l has a chance to reply to its query, the uds: sendto failed: Connection refused message is output. This is not an interesting message. Also, don't output the

Re: [Linuxptp-devel] [PATCH v2 06/27] Include TLV in replies to management commands

2014-04-05 Thread Richard Cochran
On Mon, Mar 24, 2014 at 09:53:23AM +0100, Jiri Benc wrote: The standard requires management TLV in replies to commands: An acknowledge management message is a response to a command management message. The value of the managementId shall be identical to that in the command

Re: [Linuxptp-devel] [PATCH v2 10/27] Custom management TLV PORT_ENUMERATION_NP

2014-04-05 Thread Richard Cochran
On Mon, Mar 24, 2014 at 09:53:27AM +0100, Jiri Benc wrote: Used to enumerate all ports. With the future dynamic port adding/removal, there may be holes in the port number sequence. For now, just fill it with the sequence numbers. Signed-off-by: Jiri Benc jb...@redhat.com --- clock.c |

Re: [Linuxptp-devel] [PATCH v2 11/27] Event notification: port addition/removal

2014-04-05 Thread Richard Cochran
On Mon, Mar 24, 2014 at 09:53:28AM +0100, Jiri Benc wrote: Add an event number for port adding/removal. As currently the ports are static, this event is not emitted for now. Then this patch should wait unit dynamic ports are ready. Thanks, Richard

Re: [Linuxptp-devel] [PATCH v2 12/27] Custom management TLV PORT_PROPERTIES_NP

2014-04-05 Thread Richard Cochran
On Mon, Mar 24, 2014 at 09:53:29AM +0100, Jiri Benc wrote: Will be used by phc2sys to find out interfaces corresponding to ports. This looks okay to me. Same nits as before... Signed-off-by: Jiri Benc jb...@redhat.com --- clock.c |9 + port.c | 16 tlv.c

Re: [Linuxptp-devel] [PATCH v2 26/27] Subscription time limit

2014-04-05 Thread Richard Cochran
On Mon, Mar 24, 2014 at 09:53:43AM +0100, Jiri Benc wrote: Require subscriptions to be renewed regularly. This way, the subscription automatically times out when phc2sys is killed. Signed-off-by: Jiri Benc jb...@redhat.com --- clock.c | 24 phc2sys.c | 10

Re: [Linuxptp-devel] [PATCH v2 26/27] Subscription time limit

2014-04-05 Thread Richard Cochran
On Mon, Mar 24, 2014 at 09:53:43AM +0100, Jiri Benc wrote: Require subscriptions to be renewed regularly. This way, the subscription automatically times out when phc2sys is killed. Can't this patch go before all of the phc2sys stuff? Thanks, Richard

Re: [Linuxptp-devel] [PATCH v2 10/27] Custom management TLV PORT_ENUMERATION_NP

2014-04-07 Thread Richard Cochran
On Mon, Apr 07, 2014 at 06:02:13PM +0200, Jiri Benc wrote: On Sat, 5 Apr 2014 13:02:14 +0200, Richard Cochran wrote: +struct port_enumeration_np { + UInteger16 numberPorts; + UInteger16 portNumber[0]; +} PACKED; + I don't like this variable sized message. We should stick

Re: [Linuxptp-devel] [PATCH] raw: fix reading of uninitialized memory on recv

2014-04-10 Thread Richard Cochran
On Thu, Apr 10, 2014 at 11:37:22AM +0200, Jiri Benc wrote: When less bytes than the header size is read, do not indicate to the caller that the read was successful, as the caller would read uninitialized memory. To achieve that, subtract the header size unconditionally (unless an error was

Re: [Linuxptp-devel] [PATCH 4/5] Common type holding an address

2014-04-11 Thread Richard Cochran
On Fri, Apr 11, 2014 at 12:25:54PM +0200, Jiri Benc wrote: This modifies all transports to use a new common address type, struct address. This address is stored in a ptp_message for all received messages. For sending, the default address is used with the default sending functions,

Re: [Linuxptp-devel] [PATCH 4/5] Common type holding an address

2014-04-11 Thread Richard Cochran
On Fri, Apr 11, 2014 at 06:13:22PM +0200, Jiri Benc wrote: It's the first member of the union. I originally had only struct sockaddr_storage here but it led to horrible typecasting in the code (which is quite error prone when you accidentally put an extra before the variable). In the end, I

Re: [Linuxptp-devel] kernel crashed when run linux ptp

2014-04-14 Thread Richard Cochran
On Mon, Apr 14, 2014 at 03:56:18PM +0800, Arnold kang wrote: Dear Richard, i'm sorry for i need your help again, i just resolve the kernel crash, this was caused by call spin_lock twice in stmmac mac driver, but another problem come out. when i run ./ptp4l -i eth0 -H -p /dev/ptp0

Re: [Linuxptp-devel] [PATCH 0/5] Common address struct, sendto implementation

2014-04-14 Thread Richard Cochran
FYI, I am inclined to merge this, but I am going to review it once more, in the next couple of days. Thanks, Richard -- Learn Graph Databases - Download FREE O'Reilly Book Graph Databases is the definitive new guide to

Re: [Linuxptp-devel] kernel crashed when run linux ptp

2014-04-14 Thread Richard Cochran
On Mon, Apr 14, 2014 at 05:45:34PM +0800, Arnold kang wrote: yes, i just set never timeout, but it doesn't work. But there is no such setting. ??? Richard -- Learn Graph Databases - Download FREE O'Reilly Book Graph

Re: [Linuxptp-devel] [PATCH 1/5] raw: replace hard coded constants by MAC_LEN

2014-04-16 Thread Richard Cochran
On Fri, Apr 11, 2014 at 12:25:51PM +0200, Jiri Benc wrote: Signed-off-by: Jiri Benc jb...@redhat.com --- Applied. Thanks, Richard -- Learn Graph Databases - Download FREE O'Reilly Book Graph Databases is the

Re: [Linuxptp-devel] [PATCH 2/5] raw: separate src and dst addresses

2014-04-16 Thread Richard Cochran
On Fri, Apr 11, 2014 at 12:25:52PM +0200, Jiri Benc wrote: In order to be able to convert to a generic address struct, separate source and destination address into separate fields. Signed-off-by: Jiri Benc jb...@redhat.com --- Applied. Thanks, Richard

Re: [Linuxptp-devel] [PATCH 3/5] Let transport_recv/send/peer use ptp_message

2014-04-16 Thread Richard Cochran
On Fri, Apr 11, 2014 at 12:25:53PM +0200, Jiri Benc wrote: The callers of those functions are all using ptp_message. As we're going to return more information (the address), let those functions just fill in the ptp_message fields directly. Some minor reshuffling needed to prevent circular

Re: [Linuxptp-devel] [PATCH 4/5] Common type holding an address

2014-04-16 Thread Richard Cochran
Jiri, I think this series brings a nice improvement, on second look. But the change in this patch could be better yet? See below... On Fri, Apr 11, 2014 at 12:25:54PM +0200, Jiri Benc wrote: diff --git a/transport.c b/transport.c index cb799a68aa9f..25d569607f71 100644 --- a/transport.c +++

Re: [Linuxptp-devel] what is the function rx_timestamp_work used for?

2014-04-26 Thread Richard Cochran
On Sat, Apr 26, 2014 at 02:06:08PM +0800, Arnold kang wrote: Dear Developers, can some one tell me whia't rx_timestamp_work is used for in dp83640.c, thanks! This is a Linux work item, which is a kind of delayed execution. Its purpose is to match PTP packets with their time stamps and

Re: [Linuxptp-devel] what is the function rx_timestamp_work used for?

2014-04-26 Thread Richard Cochran
On Sat, Apr 26, 2014 at 04:20:28PM +0800, Arnold kang wrote: Dear Richard, and dp83640_txtstamp call this function, the only thing do is Clear out expired time stamps, right? Yes, probably. I don't remember why the transmit function schedules the work. It might be left over from a previous

Re: [Linuxptp-devel] [PATCH v2 0/2] Common address struct, sendto implementation

2014-04-26 Thread Richard Cochran
On Tue, Apr 22, 2014 at 04:00:58PM +0200, Jiri Benc wrote: Changes in v2: removed default_addr, passing NULL address to transport send methods instead. Jiri Benc (2): Common type holding an address Implement transport_sendto Both patches applied. Thanks, Richard

Re: [Linuxptp-devel] [PATCH v3 02/11] Respond with an error to management messages to non-existing ports

2014-05-03 Thread Richard Cochran
On Fri, May 02, 2014 at 12:37:45PM +0200, Jiri Benc wrote: Signed-off-by: Jiri Benc jb...@redhat.com --- Applied. Thanks, Richard -- Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run

Re: [Linuxptp-devel] [PATCH v3 03/11] Remove unneeded parameter in port_forward

2014-05-03 Thread Richard Cochran
On Fri, May 02, 2014 at 12:37:46PM +0200, Jiri Benc wrote: This parameter was made obsolete by the common address refactoring. Signed-off-by: Jiri Benc jb...@redhat.com --- Applied. Thanks, Richard -- Accelerate

Re: [Linuxptp-devel] [PATCH v3 04/11] Implement port_forward_to

2014-05-03 Thread Richard Cochran
On Fri, May 02, 2014 at 12:37:47PM +0200, Jiri Benc wrote: Signed-off-by: Jiri Benc jb...@redhat.com --- Applied. Thanks, Richard -- Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run

Re: [Linuxptp-devel] [PATCH v3 06/11] Include TLV in replies to management commands

2014-05-03 Thread Richard Cochran
On Fri, May 02, 2014 at 12:37:49PM +0200, Jiri Benc wrote: The standard requires management TLV in replies to commands: An acknowledge management message is a response to a command management message. The value of the managementId shall be identical to that in the command

Re: [Linuxptp-devel] [PATCH v3 07/11] Subscription time limit

2014-05-03 Thread Richard Cochran
On Fri, May 02, 2014 at 12:37:50PM +0200, Jiri Benc wrote: Add expiration time to subscriptions; they need to be renewed before they expiry. This way, the subscription automatically times out when phc2sys is killed. Signed-off-by: Jiri Benc jb...@redhat.com --- clock.c | 27

Re: [Linuxptp-devel] [PATCH v3 08/11] port: event notification

2014-05-03 Thread Richard Cochran
On Fri, May 02, 2014 at 12:37:51PM +0200, Jiri Benc wrote: Split management message creation to more fine-grained functions to allow notification messages to be created. I must admit I didn't understand what you did, at first reading. Can you lose a few words about how you split the code and

Re: [Linuxptp-devel] [PATCH v3 05/11] Event subscribing

2014-05-05 Thread Richard Cochran
On Mon, May 05, 2014 at 02:39:58PM +0200, Jiri Benc wrote: I'll wrap them but the result will look uglier than this. You version looks like this: msg-management.targetPortIdentity.clockIdentity = s-targetPortIdentity.clockIdentity;

Re: [Linuxptp-devel] clock_adjtime not work in arm linux 3.0

2014-05-05 Thread Richard Cochran
On Tue, May 06, 2014 at 10:56:22AM +0800, Arnold kang wrote: Dear Richard, i'm sorry ask for your help , clock_adjtime may not work in arm linux v3.0, this syscall return success, but can not change time nor get time, anything wrong, or need add a patch ? I think you told us before that

Re: [Linuxptp-devel] [PATCH v4 0/6] automatic phc2sys configuration, ptp4l part

2014-05-06 Thread Richard Cochran
On Tue, May 06, 2014 at 06:41:45PM +0200, Jiri Benc wrote: Jiri Benc (6): Event subscribing Subscription time limit port: event notification clock: event notification Event notification: port state Custom management TLV PORT_PROPERTIES_NP Series applied. Thanks, Richard

Re: [Linuxptp-devel] [PATCH 00/14] automatic phc2sys configuration, phc2sys part

2014-05-12 Thread Richard Cochran
On Wed, May 07, 2014 at 06:23:04PM +0200, Jiri Benc wrote: phc2sys.8| 115 +-- phc2sys.c| 1076 -- pmc_common.c | 12 +- pmc_common.h |2 + 4 files changed, 908 insertions(+), 297 deletions(-) This series only

Re: [Linuxptp-devel] [PATCH] hwtstamp_ctl: add support for SIOCGHWTSTAMP

2014-05-23 Thread Richard Cochran
Nitpicking... On Wed, May 14, 2014 at 04:46:23PM -0700, Jacob Keller wrote: - while (EOF != (c = getopt(argc, argv, hi:r:t:v))) { + while (EOF != (c = getopt(argc, argv, hi:r:t:vg))) { Options in alphabetical order here ... switch (c) { case 'i':

Re: [Linuxptp-devel] [PATCH 1/2] missing: add SIOCGHWTSTAMP to missing.h

2014-06-02 Thread Richard Cochran
Never mind, this is a socket option, not an ioctl. My bad. Thanks, Richard -- Learn Graph Databases - Download FREE O'Reilly Book Graph Databases is the definitive new guide to graph databases and their applications.

Re: [Linuxptp-devel] linuxptp in kernel 2.6.30+?

2014-06-02 Thread Richard Cochran
On Mon, Jun 02, 2014 at 04:07:38PM +, Daniel Le wrote: I understand that linuxptp code base as is requires Linux kernel version 3.0 or newer. For platforms where kernel 3.x is not yet available, is it possible to backport linuxptp into a kernel 2.6.30+? (Everything is possible.) If so,

Re: [Linuxptp-devel] [PATCH 2/2] hwtstamp_ctl: use SIOCGHWTSTAMP ioctl before destructively setting policy

2014-06-03 Thread Richard Cochran
On Mon, Jun 02, 2014 at 08:59:27PM +, Keller, Jacob E wrote: Too long of a line? I usually try to follow the Linux practice which allows long lines if they only contain print statement, so that it's easier to grep the source for where the error came from.. Linux has way more lines of

Re: [Linuxptp-devel] linuxptp in kernel 2.6.30+?

2014-06-03 Thread Richard Cochran
On Mon, Jun 02, 2014 at 09:02:16PM +, Keller, Jacob E wrote: Redhat did a semi-decent job in their RHEL 6.4, and 6.5 kernels. I've tested ptp4l against them, and it mostly works. Wasn't someone asking about the RH backport on the linuxptp lists a while back, with some unresolved problem?

Re: [Linuxptp-devel] linuxptp in kernel 2.6.30+?

2014-06-03 Thread Richard Cochran
On Tue, Jun 03, 2014 at 09:53:51AM +0200, Jiri Benc wrote: I would be very interested in such problems. RHEL 6.4 support was buggy (and it was not unexpected, we marked it as Tech Preview for a reason) IIRC, the earlier post was about RHEL 6.4. The reason I mentioned the RH and TI back ports

Re: [Linuxptp-devel] linuxptp in kernel 2.6.30+?

2014-06-04 Thread Richard Cochran
On Tue, Jun 03, 2014 at 07:25:29PM +, Daniel Le wrote: I browsed both the linuxptp devel and user mail archives, but couldn't see any email thread about RHEL back porting of linuxptp and ethtool. Perhaps I didn't look at the right places or missed it. Does someone know where the relevant

Re: [Linuxptp-devel] [PATCH 1/2] missing: add SIOCGHWTSTAMP to missing.h

2014-06-04 Thread Richard Cochran
On Tue, May 27, 2014 at 10:34:35AM -0700, Jacob Keller wrote: In order to allow hwtstamp_ctl to use the non-destructive SIOCGHWTSTAMP ioctl, we need to add it to missing.h, in order to prevent build failure on older kernels which don't have this support. Signed-off-by: Jacob Keller

Re: [Linuxptp-devel] [PATCH v4] hwtstamp_ctl: use SIOCGHWTSTAMP ioctl before destructively setting policy

2014-06-04 Thread Richard Cochran
On Tue, Jun 03, 2014 at 02:03:11PM -0700, Jacob Keller wrote: This patch modifies the hwtstamp_ctl program, so that it will (attempt to) use the SIOCGHWTSTAMP ioctl to non-destructively read the current hardware timestamping policy, prior to setting it with SIOCSHWTSTAMP. Applied. Thanks,

Re: [Linuxptp-devel] timestamp types

2014-06-06 Thread Richard Cochran
On Fri, Jun 06, 2014 at 02:54:17PM +0530, Anil kumar wrote: Hi, Cany any one please let me know the difference between timestamp_types TS_HARDWARE and TS_LEGACY_HW.In the code, SOF_TIMESTAMPING_RAW_HARDWARE flag is used for TS_HARDWARE type and SOF_TIMESTAMPING_SYS_HARDWARE flag is

Re: [Linuxptp-devel] linuxptp in kernel 2.6.30+?

2014-06-09 Thread Richard Cochran
On Mon, Jun 09, 2014 at 03:53:37PM +, Daniel Le wrote: Meanwhile, I might get started with software timestamping in a Linux kernel version 2.6.30+ and just wanted to double check that I don't have to back port the PHC and can safely ignore the phc2sys program in this case? When ptp4l

Re: [Linuxptp-devel] addition of a get frequency op?

2014-06-09 Thread Richard Cochran
On Mon, Jun 09, 2014 at 10:10:56PM +, Keller, Jacob E wrote: Can this be done on the ptp clock device? If so, this would be fine :) Yes, using the FD_TO_CLOCKID macro. See phc_open() in phc.c. Thanks, Richard --

Re: [Linuxptp-devel] PTP Telecom Profile (ITU-T G.8265)

2014-06-10 Thread Richard Cochran
On Tue, Jun 10, 2014 at 06:51:31PM +, Daniel Le wrote: Hi Richard, I'm not yet familiar enough with IEEE 802.1AS-2011. From what I know, there are two kinds of AVB devices: end stations and bridges. Since linuxptp specifies support in the role of end station, does it imply

Re: [Linuxptp-devel] PTP Telecom Profile (ITU-T G.8265)

2014-06-10 Thread Richard Cochran
On Tue, Jun 10, 2014 at 07:14:42AM +0200, Richard Cochran wrote: On Mon, Jun 09, 2014 at 03:05:25PM +, Daniel Le wrote: It has been almost a year. Is there a new plan or intend to add support for this telecom profile ITU-T G.8265 and others in the future? - IEEE-C37.238 Power

Re: [Linuxptp-devel] [PATCH RFC] Add new servo for NTP SHM reference clock.

2014-06-12 Thread Richard Cochran
On Wed, Jun 11, 2014 at 06:07:08PM +0200, Miroslav Lichvar wrote: This is mainly useful when other time sources are available on the system (e.g. NTP, hardware reference clocks, or other PTP domains) and a fallback to/from PTP is needed. So, this would let ntpd mix PTP and NTP sources

Re: [Linuxptp-devel] [PATCH v3 00/15] automatic phc2sys configuration, phc2sys part

2014-06-14 Thread Richard Cochran
On Thu, Jun 12, 2014 at 11:45:22AM +0200, Miroslav Lichvar wrote: I'm happy with the set now, in my tests all seems to be working well. Series applied. Nice work, Jiri. Thanks, Richard -- HPCC Systems Open Source Big

Re: [Linuxptp-devel] linuxptp on ubuntu 14.04

2014-06-16 Thread Richard Cochran
On Mon, Jun 16, 2014 at 12:10:50PM +0100, Andrei Perietanu wrote: Hi all, I'm trying to get ptp working on my machine (running ubuntu 14.04). I followed the instructions on the website but must have missed something: when running *sudo ptp4l -i eth0 -m*I get the following messages:

Re: [Linuxptp-devel] linuxptp on ubuntu 14.04

2014-06-18 Thread Richard Cochran
On Wed, Jun 18, 2014 at 12:37:46PM +0100, Andrei Perietanu wrote: I'm running wireshark on the master Please try it on the slave, to see if the Sync messages are arriving. Thanks, Richard -- HPCC Systems Open Source

Re: [Linuxptp-devel] [PATCH 3/3] makefile: remove programs on clean.

2014-06-18 Thread Richard Cochran
On Wed, Jun 18, 2014 at 03:44:50PM +0200, Miroslav Lichvar wrote: clean: - rm -f $(OBJECTS) $(DEPEND) + rm -f $(OBJECTS) $(DEPEND) $(PRG) distclean: clean rm -f $(PRG) The clean target removes all the build products except for the executables. The distclean target goes one

Re: [Linuxptp-devel] linuxptp on ubuntu 14.04

2014-06-18 Thread Richard Cochran
On Wed, Jun 18, 2014 at 01:13:27PM +0100, Andrei Perietanu wrote: I tried running wireshark on both nodes and I see the exact same wireshark output on both of them, so sync messages are being received by the slave. Okay, then something very strange is happening, and I can't imagine what is

Re: [Linuxptp-devel] Conflict in DOMAIN with math.h

2014-07-08 Thread Richard Cochran
On Wed, Jun 25, 2014 at 12:57:59PM +0200, Miroslav Lichvar wrote: On Wed, Jun 25, 2014 at 12:50:15PM +0200, Miroslav Lichvar wrote: Ok. I wasn't sure to what would be a good name instead of DOMAIN and it seems some other names like TIME could be problematic as well, so I prefixed all of

Re: [Linuxptp-devel] PTP+NTP configuration

2014-07-17 Thread Richard Cochran
On Thu, Jul 17, 2014 at 03:15:20PM +0200, Miroslav Lichvar wrote: Would you be interested in including this program in linuxptp? Yes. Are dependencies on external programs acceptable? That depends. Will the library be available on every system that wants to run linuxptp? I spend a lot of

Re: [Linuxptp-devel] PTP+NTP configuration

2014-07-18 Thread Richard Cochran
On Fri, Jul 18, 2014 at 11:08:15AM +0200, Miroslav Lichvar wrote: That's why I'm asking. The new program needs at least ntpd or chronyd installed on the system to function properly. Are you ok with having code in linuxptp that's not actually related to PTP? Run time dependencies like that

Re: [Linuxptp-devel] 1588 PTP implementation

2014-07-18 Thread Richard Cochran
On Fri, Jul 11, 2014 at 02:03:03AM +0800, Ronex Dicapriyo wrote: Hello, I need to implement functionality of 1588 PTP standard in some harware IP model. For this I have few queries as below: 1) Linux ptp implementation - How it uses/works for harware and software timestamping ? In

Re: [Linuxptp-devel] [PATCH v4] linuxptp: add phc_ctl program to help debug PHC devices

2014-07-18 Thread Richard Cochran
Jacob, I tried this program out, and ... On Thu, Jul 10, 2014 at 03:49:30PM -0700, Jacob Keller wrote: +.BI caps +Display the device capabiltiies. This is the default command if no commands are +provided. With no command, nothing is displayed. +Quickly sanity check frequency slewing by

Re: [Linuxptp-devel] [PATCH v4 6/6] Dynamic allocation of interface config entries

2014-07-19 Thread Richard Cochran
On Fri, Jun 27, 2014 at 07:28:17PM +0200, Jiri Benc wrote: -/* returns the number matching that interface, or -1 on failure */ -int config_create_interface(char *name, struct config *cfg) +struct interface *config_create_interface(char *name, struct config *cfg) { struct interface

Re: [Linuxptp-devel] [PATCH 0/5] Allow running multiple pmc or phc2sys instances

2014-09-21 Thread Richard Cochran
On Tue, Jul 08, 2014 at 04:14:17PM +0200, Miroslav Lichvar wrote: The first patch allows using phc2sys with non-default server UDS path. The next three patches add signal handling to pmc/phc2sys and remove the UDS socket before exit. The last patch appends the process ID to the local UDS path

Re: [Linuxptp-devel] [PATCH 0/3] timemaster

2014-09-21 Thread Richard Cochran
This has some warnings, /home/richard/git/linuxptp/util.c: In function ‘string_newf’: /home/richard/git/linuxptp/util.c:353:2: error: implicit declaration of function ‘vasprintf’ [-Werror=implicit-function-declaration] /home/richard/git/linuxptp/timemaster.c: In function 'start_program':

Re: [Linuxptp-devel] [PATCH] Fix Coverity warning in sk_interface_addr().

2014-09-21 Thread Richard Cochran
On Wed, Sep 17, 2014 at 11:11:16AM +0200, Miroslav Lichvar wrote: Copy the address directly to struct sockaddr_in or sockaddr_in6 instead of sockaddr as Coverity doesn't seem to understand the union and reports a buffer overflow. Signed-off-by: Miroslav Lichvar mlich...@redhat.com Applied.

Re: [Linuxptp-devel] [PATCH v5 0/6] Dynamic port allocation

2014-09-21 Thread Richard Cochran
On Thu, Aug 14, 2014 at 03:56:00PM +0200, Jiri Benc wrote: Currently, there's a limit of MAX_PORTS ports (8 by default). This patchset removes this limitation by using linked lists instead of fixed size arrays. The core patches are 3/6 and 6/6, the rest are prerequisites: patch 1/6 is moving

Re: [Linuxptp-devel] [PATCH 0/3] timemaster

2014-09-25 Thread Richard Cochran
On Thu, Sep 25, 2014 at 05:21:39PM +0200, Miroslav Lichvar wrote: On Sun, Sep 21, 2014 at 12:57:33PM +0200, Richard Cochran wrote: This has some warnings, /home/richard/git/linuxptp/util.c: In function ‘string_newf’: /home/richard/git/linuxptp/util.c:353:2: error: implicit declaration

Re: [Linuxptp-devel] [PATCH 0/3] timemaster

2014-09-30 Thread Richard Cochran
On Tue, Sep 30, 2014 at 04:21:48PM +0200, Miroslav Lichvar wrote: On Thu, Sep 25, 2014 at 06:53:32PM +0200, Richard Cochran wrote: On Thu, Sep 25, 2014 at 05:21:39PM +0200, Miroslav Lichvar wrote: Is libc on your system missing vasprintf() or it just needs _GNU_SOURCE to be defined

Re: [Linuxptp-devel] [PATCH 0/3] timemaster

2014-10-01 Thread Richard Cochran
On Wed, Oct 01, 2014 at 12:22:36PM +0200, Miroslav Lichvar wrote: Hm, I'd like to see all binaries build by default on normal systems that have fork. Ok. How about using posix_spawn() instead? It seems uclibc includes support for it, so we could build timemaster even there (although I'm

Re: [Linuxptp-devel] [PATCH 4/4] Add timemaster. (compressed)

2014-10-02 Thread Richard Cochran
On Thu, Oct 02, 2014 at 01:59:39PM +0200, Miroslav Lichvar wrote: This was larger than then 40KB list limit, sending again as a compressed attachment. Sorry about that. I bumped the limit up to 100k now in mailman. Thanks, Richard

Re: [Linuxptp-devel] [PATCH 3/4] Add string and pointer array utility functions.

2014-10-03 Thread Richard Cochran
On Thu, Oct 02, 2014 at 10:38:33AM +0200, Miroslav Lichvar wrote: Add some functions to work with strings and arrays of pointers that will be useful later. Signed-off-by: Miroslav Lichvar mlich...@redhat.com --- incdefs.sh | 3 ++ util.c | 95

Re: [Linuxptp-devel] [PATCHv3 0/4] timemaster

2014-10-03 Thread Richard Cochran
On Fri, Oct 03, 2014 at 02:13:48PM +0200, Miroslav Lichvar wrote: Changes since v2: - fix warnings reported by Clang static analyzer - don't define _GNU_SOURCE twice Changes since v1: - add support for posix_spawn() to compile on systems missing fork() - define _GNU_SOURCE for vasprintf()

Re: [Linuxptp-devel] [PATCH] sk: Modify poll to poll on ERRQUEUE

2014-10-31 Thread Richard Cochran
On Fri, Oct 31, 2014 at 02:05:14PM -0500, Joe Schaack wrote: Implement the SO_SELECT_ERR_QUEUE socket option that enables ERRQUEUE messages to be polled for under the POLLPRI flag, greatly increasing the number of packets per second that can be sent from linuxptp. SO_SELECT_ERR_QUEUE is only

[Linuxptp-devel] [PATCH RFC 1/3] config: add a option to enable a poor man's boundary clock.

2014-11-01 Thread Richard Cochran
-off-by: Richard Cochran richardcoch...@gmail.com --- config.c| 13 + config.h|1 + default.cfg |1 + ds.h|1 + gPTP.cfg|1 + ptp4l.8 | 12 +++- ptp4l.c |1 + 7 files changed, 29 insertions(+), 1 deletion(-) diff --git

[Linuxptp-devel] [PATCH RFC 0/3] Poor man's boundary clock

2014-11-01 Thread Richard Cochran
version. Feedback is most welcome. Thanks, Richard Richard Cochran (3): config: add a option to enable a poor man's boundary clock. port: allow running a boundary clock with multiple clock devices. phc2sys: make automatic mode actually work. config.c| 13 + config.h

[Linuxptp-devel] [PATCH RFC 3/3] phc2sys: make automatic mode actually work.

2014-11-01 Thread Richard Cochran
will not synchronize it, because the SLAVE port's state variable will have been cleared to zero, effectively erasing the local time source. This patch fixes the issue by removing the assignment. Signed-off-by: Richard Cochran richardcoch...@gmail.com --- phc2sys.c |1 - 1 file changed, 1

[Linuxptp-devel] [PATCH RFC 2/3] port: allow running a boundary clock with multiple clock devices.

2014-11-01 Thread Richard Cochran
If the user has configured the appropriate option, then simply warn about the clock device mismatch, and then go on. Signed-off-by: Richard Cochran richardcoch...@gmail.com --- port.c |6 +- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/port.c b/port.c index f39ad1e

Re: [Linuxptp-devel] [PATCH RFC 0/3] Poor man's boundary clock

2014-11-05 Thread Richard Cochran
On Wed, Nov 05, 2014 at 05:56:28PM +0100, Jiri Benc wrote: My plan for the next steps has been allowing ptp4l to work with multiple independent PHCs that would form a PTP clock (and rely on phc2sys to sync those PHCs). Doesn't my patch #2 do this? This needs separation of struct clock into

Re: [Linuxptp-devel] [PATCH RFC 0/3] Poor man's boundary clock

2014-11-05 Thread Richard Cochran
On Wed, Nov 05, 2014 at 08:28:04PM +0100, Jiri Benc wrote: I'm not sure whether this patch was enough to support the boundary clock or more was needed but I remember I had the boundary clock stuff done (though untested) and I cannot find anything on top of this, so this was probably enough.

Re: [Linuxptp-devel] [PATCH] port: fix fda initialization.

2014-11-05 Thread Richard Cochran
On Wed, Nov 05, 2014 at 12:22:43PM +0100, Miroslav Lichvar wrote: Initialize all descriptors to prevent polling invalid descriptors from faulty ports. Signed-off-by: Miroslav Lichvar mlich...@redhat.com --- port.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git

[Linuxptp-devel] [PATCH RFC V2 06/10] config: add a option to enable a poor man's boundary clock.

2014-11-05 Thread Richard Cochran
-off-by: Richard Cochran richardcoch...@gmail.com --- config.c| 13 + config.h|1 + default.cfg |1 + ds.h|1 + gPTP.cfg|1 + ptp4l.8 | 12 +++- ptp4l.c |1 + 7 files changed, 29 insertions(+), 1 deletion(-) diff --git

[Linuxptp-devel] [PATCH RFC V2 03/10] trivial: do not assign a FP constant to an integer.

2014-11-05 Thread Richard Cochran
Signed-off-by: Richard Cochran richardcoch...@gmail.com --- clock.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/clock.c b/clock.c index 726999e..df82d2b 100644 --- a/clock.c +++ b/clock.c @@ -789,7 +789,7 @@ struct clock *clock_create(int phc_index, struct

[Linuxptp-devel] [PATCH RFC V2 07/10] phc2sys: make automatic mode actually work.

2014-11-05 Thread Richard Cochran
will not synchronize it, because the SLAVE port's state variable will have been cleared to zero, effectively erasing the local time source. This patch fixes the issue by removing the assignment. Signed-off-by: Richard Cochran richardcoch...@gmail.com --- phc2sys.c |1 - 1 file changed, 1

[Linuxptp-devel] [PATCH RFC V2 08/10] phc2sys: automatic mode: synchronize all non-slave ports.

2014-11-05 Thread Richard Cochran
When running a jbod Boundary Clock, as long as we have one slaved port, we always want the clocks on the other ports to be synchronized, regardless of their port state. Signed-off-by: Richard Cochran richardcoch...@gmail.com --- phc2sys.c | 40 +--- 1 file

[Linuxptp-devel] [PATCH RFC V2 01/10] Use SO_SELECT_ERR_QUEUE when available.

2014-11-05 Thread Richard Cochran
if necessary. Suggested-by: Joe Schaack jscha...@xes-inc.com Signed-off-by: Richard Cochran richardcoch...@gmail.com --- missing.h |4 sk.c | 16 ++-- 2 files changed, 18 insertions(+), 2 deletions(-) diff --git a/missing.h b/missing.h index 43ac6cf..f7efd92 100644

[Linuxptp-devel] [PATCH RFC V2 09/10] Push the node's clock ID into the port data structure.

2014-11-05 Thread Richard Cochran
This patch lets the ports (rather than the clock) remember the clock ID, passing it back to the clock code during synchronization. This does not represent a functional change, but rather paves the way for ports to override the clock ID when running a jbod BC. Signed-off-by: Richard Cochran

Re: [Linuxptp-devel] [PATCHv2] port: fix fda initialization.

2014-11-07 Thread Richard Cochran
On Thu, Nov 06, 2014 at 09:52:49AM +0100, Miroslav Lichvar wrote: Initialize all descriptors to prevent polling invalid descriptors from faulty ports. Signed-off-by: Miroslav Lichvar mlich...@redhat.com --- port.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) Applied.

[Linuxptp-devel] [PATCH RFC V3 1/5] config: add a option to enable a poor man's boundary clock.

2014-11-21 Thread Richard Cochran
-off-by: Richard Cochran richardcoch...@gmail.com Acked-by: Jiri Benc jb...@redhat.com --- config.c| 13 + config.h|1 + default.cfg |1 + ds.h|1 + gPTP.cfg|1 + ptp4l.8 | 12 +++- ptp4l.c |1 + 7 files changed, 29 insertions

[Linuxptp-devel] [PATCH RFC V3 3/5] Push the node's clock ID into the port data structure.

2014-11-21 Thread Richard Cochran
This patch lets the ports (rather than the clock) remember the clock ID, passing it back to the clock code during synchronization. This does not represent a functional change, but rather paves the way for ports to override the clock ID when running a jbod BC. Signed-off-by: Richard Cochran

[Linuxptp-devel] [PATCH RFC V3 0/5] Poor man's boundary clock

2014-11-21 Thread Richard Cochran
JBOD BC start up messages - Synchronize the group of devices, even if none is a slave Richard Cochran (5): config: add a option to enable a poor man's boundary clock. phc2sys: automatic mode: synchronize all non-slave ports. Push the node's clock ID into the port data structure. port: allow

  1   2   3   4   5   6   7   8   9   10   >